Training in progress, step 280, checkpoint
Browse files
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 80013120
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9d19d032b2720552ed5a8c04c8453d710ed0eed172ae313734cb428d3f003fc
|
| 3 |
size 80013120
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08e5683a29463e32746f14f186f042dd447b12cafcad678bbbddb34b9249098a
|
| 3 |
+
size 41120084
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3369e2942ff752b68da734b9eaf1a12b8c42e1d8b80214950313c71f22a426be
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9fe9c01b8c53647998de80cbc88fe3102f7ee94466c3d3ba6db0d6d4b3bdc06d
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 140,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -995,6 +995,994 @@
|
|
| 995 |
"eval_samples_per_second": 17.654,
|
| 996 |
"eval_steps_per_second": 8.827,
|
| 997 |
"step": 140
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 998 |
}
|
| 999 |
],
|
| 1000 |
"logging_steps": 1,
|
|
@@ -1014,7 +2002,7 @@
|
|
| 1014 |
"attributes": {}
|
| 1015 |
}
|
| 1016 |
},
|
| 1017 |
-
"total_flos":
|
| 1018 |
"train_batch_size": 2,
|
| 1019 |
"trial_name": null,
|
| 1020 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5,
|
| 5 |
"eval_steps": 140,
|
| 6 |
+
"global_step": 280,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 995 |
"eval_samples_per_second": 17.654,
|
| 996 |
"eval_steps_per_second": 8.827,
|
| 997 |
"step": 140
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.2517857142857143,
|
| 1001 |
+
"grad_norm": 0.36249810457229614,
|
| 1002 |
+
"learning_rate": 0.00017328668381631318,
|
| 1003 |
+
"loss": 1.1208,
|
| 1004 |
+
"step": 141
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.25357142857142856,
|
| 1008 |
+
"grad_norm": 0.3750612735748291,
|
| 1009 |
+
"learning_rate": 0.00017289686274214118,
|
| 1010 |
+
"loss": 1.2502,
|
| 1011 |
+
"step": 142
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.25535714285714284,
|
| 1015 |
+
"grad_norm": 0.4201869070529938,
|
| 1016 |
+
"learning_rate": 0.0001725046632837007,
|
| 1017 |
+
"loss": 1.1947,
|
| 1018 |
+
"step": 143
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.2571428571428571,
|
| 1022 |
+
"grad_norm": 0.4865645468235016,
|
| 1023 |
+
"learning_rate": 0.00017211009823716694,
|
| 1024 |
+
"loss": 0.8749,
|
| 1025 |
+
"step": 144
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.25892857142857145,
|
| 1029 |
+
"grad_norm": 0.38693225383758545,
|
| 1030 |
+
"learning_rate": 0.00017171318047589637,
|
| 1031 |
+
"loss": 1.2495,
|
| 1032 |
+
"step": 145
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.26071428571428573,
|
| 1036 |
+
"grad_norm": 0.40707525610923767,
|
| 1037 |
+
"learning_rate": 0.00017131392295000674,
|
| 1038 |
+
"loss": 1.2321,
|
| 1039 |
+
"step": 146
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.2625,
|
| 1043 |
+
"grad_norm": 0.39570894837379456,
|
| 1044 |
+
"learning_rate": 0.00017091233868595467,
|
| 1045 |
+
"loss": 1.301,
|
| 1046 |
+
"step": 147
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.2642857142857143,
|
| 1050 |
+
"grad_norm": 0.4085226058959961,
|
| 1051 |
+
"learning_rate": 0.00017050844078611056,
|
| 1052 |
+
"loss": 1.5369,
|
| 1053 |
+
"step": 148
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.26607142857142857,
|
| 1057 |
+
"grad_norm": 0.47094810009002686,
|
| 1058 |
+
"learning_rate": 0.0001701022424283311,
|
| 1059 |
+
"loss": 1.9374,
|
| 1060 |
+
"step": 149
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.26785714285714285,
|
| 1064 |
+
"grad_norm": 0.8517308831214905,
|
| 1065 |
+
"learning_rate": 0.00016969375686552937,
|
| 1066 |
+
"loss": 1.808,
|
| 1067 |
+
"step": 150
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.26964285714285713,
|
| 1071 |
+
"grad_norm": 0.1922745406627655,
|
| 1072 |
+
"learning_rate": 0.00016928299742524234,
|
| 1073 |
+
"loss": 1.6608,
|
| 1074 |
+
"step": 151
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.2714285714285714,
|
| 1078 |
+
"grad_norm": 0.2090916484594345,
|
| 1079 |
+
"learning_rate": 0.00016886997750919619,
|
| 1080 |
+
"loss": 1.8009,
|
| 1081 |
+
"step": 152
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.2732142857142857,
|
| 1085 |
+
"grad_norm": 0.21698515117168427,
|
| 1086 |
+
"learning_rate": 0.00016845471059286887,
|
| 1087 |
+
"loss": 1.7821,
|
| 1088 |
+
"step": 153
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.275,
|
| 1092 |
+
"grad_norm": 0.21791532635688782,
|
| 1093 |
+
"learning_rate": 0.00016803721022505067,
|
| 1094 |
+
"loss": 1.5901,
|
| 1095 |
+
"step": 154
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.2767857142857143,
|
| 1099 |
+
"grad_norm": 0.22199980914592743,
|
| 1100 |
+
"learning_rate": 0.00016761749002740193,
|
| 1101 |
+
"loss": 1.7047,
|
| 1102 |
+
"step": 155
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.2785714285714286,
|
| 1106 |
+
"grad_norm": 0.2096625566482544,
|
| 1107 |
+
"learning_rate": 0.0001671955636940088,
|
| 1108 |
+
"loss": 1.6898,
|
| 1109 |
+
"step": 156
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.28035714285714286,
|
| 1113 |
+
"grad_norm": 0.22975414991378784,
|
| 1114 |
+
"learning_rate": 0.00016677144499093626,
|
| 1115 |
+
"loss": 1.7631,
|
| 1116 |
+
"step": 157
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.28214285714285714,
|
| 1120 |
+
"grad_norm": 0.2187148928642273,
|
| 1121 |
+
"learning_rate": 0.0001663451477557792,
|
| 1122 |
+
"loss": 1.7872,
|
| 1123 |
+
"step": 158
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.2839285714285714,
|
| 1127 |
+
"grad_norm": 0.2257414609193802,
|
| 1128 |
+
"learning_rate": 0.0001659166858972107,
|
| 1129 |
+
"loss": 1.7732,
|
| 1130 |
+
"step": 159
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.2857142857142857,
|
| 1134 |
+
"grad_norm": 0.22986693680286407,
|
| 1135 |
+
"learning_rate": 0.00016548607339452853,
|
| 1136 |
+
"loss": 1.7031,
|
| 1137 |
+
"step": 160
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.2875,
|
| 1141 |
+
"grad_norm": 0.21585014462471008,
|
| 1142 |
+
"learning_rate": 0.0001650533242971987,
|
| 1143 |
+
"loss": 1.8421,
|
| 1144 |
+
"step": 161
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.2892857142857143,
|
| 1148 |
+
"grad_norm": 0.22519604861736298,
|
| 1149 |
+
"learning_rate": 0.00016461845272439741,
|
| 1150 |
+
"loss": 1.6529,
|
| 1151 |
+
"step": 162
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.2910714285714286,
|
| 1155 |
+
"grad_norm": 0.22279705107212067,
|
| 1156 |
+
"learning_rate": 0.0001641814728645502,
|
| 1157 |
+
"loss": 1.9288,
|
| 1158 |
+
"step": 163
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.29285714285714287,
|
| 1162 |
+
"grad_norm": 0.22392615675926208,
|
| 1163 |
+
"learning_rate": 0.000163742398974869,
|
| 1164 |
+
"loss": 1.693,
|
| 1165 |
+
"step": 164
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.29464285714285715,
|
| 1169 |
+
"grad_norm": 0.22729454934597015,
|
| 1170 |
+
"learning_rate": 0.00016330124538088705,
|
| 1171 |
+
"loss": 1.7027,
|
| 1172 |
+
"step": 165
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.29642857142857143,
|
| 1176 |
+
"grad_norm": 0.2229882776737213,
|
| 1177 |
+
"learning_rate": 0.00016285802647599156,
|
| 1178 |
+
"loss": 1.8262,
|
| 1179 |
+
"step": 166
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.2982142857142857,
|
| 1183 |
+
"grad_norm": 0.25520074367523193,
|
| 1184 |
+
"learning_rate": 0.00016241275672095395,
|
| 1185 |
+
"loss": 1.6009,
|
| 1186 |
+
"step": 167
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.3,
|
| 1190 |
+
"grad_norm": 0.24272315204143524,
|
| 1191 |
+
"learning_rate": 0.00016196545064345812,
|
| 1192 |
+
"loss": 1.9227,
|
| 1193 |
+
"step": 168
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.30178571428571427,
|
| 1197 |
+
"grad_norm": 0.24380216002464294,
|
| 1198 |
+
"learning_rate": 0.00016151612283762652,
|
| 1199 |
+
"loss": 1.5198,
|
| 1200 |
+
"step": 169
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.30357142857142855,
|
| 1204 |
+
"grad_norm": 0.3242342472076416,
|
| 1205 |
+
"learning_rate": 0.00016106478796354382,
|
| 1206 |
+
"loss": 1.6981,
|
| 1207 |
+
"step": 170
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.3053571428571429,
|
| 1211 |
+
"grad_norm": 0.277855783700943,
|
| 1212 |
+
"learning_rate": 0.00016061146074677885,
|
| 1213 |
+
"loss": 1.7011,
|
| 1214 |
+
"step": 171
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.30714285714285716,
|
| 1218 |
+
"grad_norm": 0.2710039019584656,
|
| 1219 |
+
"learning_rate": 0.00016015615597790388,
|
| 1220 |
+
"loss": 1.7522,
|
| 1221 |
+
"step": 172
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.30892857142857144,
|
| 1225 |
+
"grad_norm": 0.26541268825531006,
|
| 1226 |
+
"learning_rate": 0.00015969888851201226,
|
| 1227 |
+
"loss": 1.3804,
|
| 1228 |
+
"step": 173
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.3107142857142857,
|
| 1232 |
+
"grad_norm": 0.28985923528671265,
|
| 1233 |
+
"learning_rate": 0.00015923967326823368,
|
| 1234 |
+
"loss": 1.6453,
|
| 1235 |
+
"step": 174
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.3125,
|
| 1239 |
+
"grad_norm": 0.33939245343208313,
|
| 1240 |
+
"learning_rate": 0.00015877852522924732,
|
| 1241 |
+
"loss": 1.1725,
|
| 1242 |
+
"step": 175
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.3142857142857143,
|
| 1246 |
+
"grad_norm": 0.29770731925964355,
|
| 1247 |
+
"learning_rate": 0.0001583154594407932,
|
| 1248 |
+
"loss": 1.6746,
|
| 1249 |
+
"step": 176
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.31607142857142856,
|
| 1253 |
+
"grad_norm": 0.3280562460422516,
|
| 1254 |
+
"learning_rate": 0.0001578504910111811,
|
| 1255 |
+
"loss": 1.1357,
|
| 1256 |
+
"step": 177
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.31785714285714284,
|
| 1260 |
+
"grad_norm": 0.2856597304344177,
|
| 1261 |
+
"learning_rate": 0.00015738363511079776,
|
| 1262 |
+
"loss": 1.1127,
|
| 1263 |
+
"step": 178
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.3196428571428571,
|
| 1267 |
+
"grad_norm": 0.316491961479187,
|
| 1268 |
+
"learning_rate": 0.00015691490697161182,
|
| 1269 |
+
"loss": 1.4281,
|
| 1270 |
+
"step": 179
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.32142857142857145,
|
| 1274 |
+
"grad_norm": 0.3632654845714569,
|
| 1275 |
+
"learning_rate": 0.00015644432188667695,
|
| 1276 |
+
"loss": 1.3413,
|
| 1277 |
+
"step": 180
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.32321428571428573,
|
| 1281 |
+
"grad_norm": 0.34329405426979065,
|
| 1282 |
+
"learning_rate": 0.00015597189520963277,
|
| 1283 |
+
"loss": 1.0579,
|
| 1284 |
+
"step": 181
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.325,
|
| 1288 |
+
"grad_norm": 0.32447105646133423,
|
| 1289 |
+
"learning_rate": 0.00015549764235420405,
|
| 1290 |
+
"loss": 1.243,
|
| 1291 |
+
"step": 182
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.3267857142857143,
|
| 1295 |
+
"grad_norm": 0.3558500409126282,
|
| 1296 |
+
"learning_rate": 0.0001550215787936977,
|
| 1297 |
+
"loss": 1.1376,
|
| 1298 |
+
"step": 183
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.32857142857142857,
|
| 1302 |
+
"grad_norm": 0.3373570740222931,
|
| 1303 |
+
"learning_rate": 0.00015454372006049803,
|
| 1304 |
+
"loss": 1.1251,
|
| 1305 |
+
"step": 184
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.33035714285714285,
|
| 1309 |
+
"grad_norm": 0.36412546038627625,
|
| 1310 |
+
"learning_rate": 0.00015406408174555976,
|
| 1311 |
+
"loss": 1.3238,
|
| 1312 |
+
"step": 185
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.33214285714285713,
|
| 1316 |
+
"grad_norm": 0.364442378282547,
|
| 1317 |
+
"learning_rate": 0.00015358267949789966,
|
| 1318 |
+
"loss": 0.9448,
|
| 1319 |
+
"step": 186
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.3339285714285714,
|
| 1323 |
+
"grad_norm": 0.3172107934951782,
|
| 1324 |
+
"learning_rate": 0.00015309952902408576,
|
| 1325 |
+
"loss": 1.2744,
|
| 1326 |
+
"step": 187
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.3357142857142857,
|
| 1330 |
+
"grad_norm": 0.34173399209976196,
|
| 1331 |
+
"learning_rate": 0.00015261464608772488,
|
| 1332 |
+
"loss": 1.0923,
|
| 1333 |
+
"step": 188
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.3375,
|
| 1337 |
+
"grad_norm": 0.33419185876846313,
|
| 1338 |
+
"learning_rate": 0.0001521280465089484,
|
| 1339 |
+
"loss": 1.2762,
|
| 1340 |
+
"step": 189
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.3392857142857143,
|
| 1344 |
+
"grad_norm": 0.3866868317127228,
|
| 1345 |
+
"learning_rate": 0.0001516397461638962,
|
| 1346 |
+
"loss": 0.9595,
|
| 1347 |
+
"step": 190
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.3410714285714286,
|
| 1351 |
+
"grad_norm": 0.3978990614414215,
|
| 1352 |
+
"learning_rate": 0.00015114976098419842,
|
| 1353 |
+
"loss": 0.9993,
|
| 1354 |
+
"step": 191
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.34285714285714286,
|
| 1358 |
+
"grad_norm": 0.3546142876148224,
|
| 1359 |
+
"learning_rate": 0.00015065810695645584,
|
| 1360 |
+
"loss": 1.3421,
|
| 1361 |
+
"step": 192
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.34464285714285714,
|
| 1365 |
+
"grad_norm": 0.39728498458862305,
|
| 1366 |
+
"learning_rate": 0.00015016480012171828,
|
| 1367 |
+
"loss": 1.1209,
|
| 1368 |
+
"step": 193
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.3464285714285714,
|
| 1372 |
+
"grad_norm": 0.4170741140842438,
|
| 1373 |
+
"learning_rate": 0.00014966985657496114,
|
| 1374 |
+
"loss": 1.0024,
|
| 1375 |
+
"step": 194
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.3482142857142857,
|
| 1379 |
+
"grad_norm": 0.4226652681827545,
|
| 1380 |
+
"learning_rate": 0.0001491732924645604,
|
| 1381 |
+
"loss": 1.3139,
|
| 1382 |
+
"step": 195
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.35,
|
| 1386 |
+
"grad_norm": 0.3712114691734314,
|
| 1387 |
+
"learning_rate": 0.00014867512399176563,
|
| 1388 |
+
"loss": 1.1574,
|
| 1389 |
+
"step": 196
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.3517857142857143,
|
| 1393 |
+
"grad_norm": 0.3655322790145874,
|
| 1394 |
+
"learning_rate": 0.00014817536741017152,
|
| 1395 |
+
"loss": 1.6149,
|
| 1396 |
+
"step": 197
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.3535714285714286,
|
| 1400 |
+
"grad_norm": 0.4362059533596039,
|
| 1401 |
+
"learning_rate": 0.0001476740390251875,
|
| 1402 |
+
"loss": 1.7657,
|
| 1403 |
+
"step": 198
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.35535714285714287,
|
| 1407 |
+
"grad_norm": 0.43134769797325134,
|
| 1408 |
+
"learning_rate": 0.00014717115519350567,
|
| 1409 |
+
"loss": 1.7167,
|
| 1410 |
+
"step": 199
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.35714285714285715,
|
| 1414 |
+
"grad_norm": 0.7784890532493591,
|
| 1415 |
+
"learning_rate": 0.00014666673232256738,
|
| 1416 |
+
"loss": 2.036,
|
| 1417 |
+
"step": 200
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.35892857142857143,
|
| 1421 |
+
"grad_norm": 0.17376984655857086,
|
| 1422 |
+
"learning_rate": 0.0001461607868700276,
|
| 1423 |
+
"loss": 1.4856,
|
| 1424 |
+
"step": 201
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.3607142857142857,
|
| 1428 |
+
"grad_norm": 0.2141953408718109,
|
| 1429 |
+
"learning_rate": 0.00014565333534321826,
|
| 1430 |
+
"loss": 1.7491,
|
| 1431 |
+
"step": 202
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.3625,
|
| 1435 |
+
"grad_norm": 0.22548137605190277,
|
| 1436 |
+
"learning_rate": 0.00014514439429860943,
|
| 1437 |
+
"loss": 1.8457,
|
| 1438 |
+
"step": 203
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.36428571428571427,
|
| 1442 |
+
"grad_norm": 0.20618294179439545,
|
| 1443 |
+
"learning_rate": 0.0001446339803412692,
|
| 1444 |
+
"loss": 1.4987,
|
| 1445 |
+
"step": 204
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.36607142857142855,
|
| 1449 |
+
"grad_norm": 0.21025151014328003,
|
| 1450 |
+
"learning_rate": 0.00014412211012432212,
|
| 1451 |
+
"loss": 1.5568,
|
| 1452 |
+
"step": 205
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.3678571428571429,
|
| 1456 |
+
"grad_norm": 0.21678180992603302,
|
| 1457 |
+
"learning_rate": 0.00014360880034840554,
|
| 1458 |
+
"loss": 1.7841,
|
| 1459 |
+
"step": 206
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.36964285714285716,
|
| 1463 |
+
"grad_norm": 0.20914790034294128,
|
| 1464 |
+
"learning_rate": 0.0001430940677611249,
|
| 1465 |
+
"loss": 1.6693,
|
| 1466 |
+
"step": 207
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.37142857142857144,
|
| 1470 |
+
"grad_norm": 0.21597585082054138,
|
| 1471 |
+
"learning_rate": 0.00014257792915650728,
|
| 1472 |
+
"loss": 1.648,
|
| 1473 |
+
"step": 208
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.3732142857142857,
|
| 1477 |
+
"grad_norm": 0.23697789013385773,
|
| 1478 |
+
"learning_rate": 0.00014206040137445348,
|
| 1479 |
+
"loss": 1.7616,
|
| 1480 |
+
"step": 209
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.375,
|
| 1484 |
+
"grad_norm": 0.2535800635814667,
|
| 1485 |
+
"learning_rate": 0.00014154150130018866,
|
| 1486 |
+
"loss": 2.0279,
|
| 1487 |
+
"step": 210
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.3767857142857143,
|
| 1491 |
+
"grad_norm": 0.21204812824726105,
|
| 1492 |
+
"learning_rate": 0.0001410212458637112,
|
| 1493 |
+
"loss": 1.8472,
|
| 1494 |
+
"step": 211
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.37857142857142856,
|
| 1498 |
+
"grad_norm": 0.36059629917144775,
|
| 1499 |
+
"learning_rate": 0.00014049965203924054,
|
| 1500 |
+
"loss": 1.8042,
|
| 1501 |
+
"step": 212
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.38035714285714284,
|
| 1505 |
+
"grad_norm": 0.21400661766529083,
|
| 1506 |
+
"learning_rate": 0.0001399767368446634,
|
| 1507 |
+
"loss": 1.698,
|
| 1508 |
+
"step": 213
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.3821428571428571,
|
| 1512 |
+
"grad_norm": 0.24055758118629456,
|
| 1513 |
+
"learning_rate": 0.00013945251734097828,
|
| 1514 |
+
"loss": 1.8758,
|
| 1515 |
+
"step": 214
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.38392857142857145,
|
| 1519 |
+
"grad_norm": 0.23605166375637054,
|
| 1520 |
+
"learning_rate": 0.00013892701063173918,
|
| 1521 |
+
"loss": 1.7425,
|
| 1522 |
+
"step": 215
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.38571428571428573,
|
| 1526 |
+
"grad_norm": 0.23343758285045624,
|
| 1527 |
+
"learning_rate": 0.00013840023386249713,
|
| 1528 |
+
"loss": 1.8683,
|
| 1529 |
+
"step": 216
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.3875,
|
| 1533 |
+
"grad_norm": 0.2475200593471527,
|
| 1534 |
+
"learning_rate": 0.00013787220422024134,
|
| 1535 |
+
"loss": 1.9091,
|
| 1536 |
+
"step": 217
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.3892857142857143,
|
| 1540 |
+
"grad_norm": 0.2618944048881531,
|
| 1541 |
+
"learning_rate": 0.00013734293893283783,
|
| 1542 |
+
"loss": 1.5086,
|
| 1543 |
+
"step": 218
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.39107142857142857,
|
| 1547 |
+
"grad_norm": 0.2627498209476471,
|
| 1548 |
+
"learning_rate": 0.00013681245526846783,
|
| 1549 |
+
"loss": 1.3878,
|
| 1550 |
+
"step": 219
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.39285714285714285,
|
| 1554 |
+
"grad_norm": 0.24390314519405365,
|
| 1555 |
+
"learning_rate": 0.0001362807705350641,
|
| 1556 |
+
"loss": 1.7332,
|
| 1557 |
+
"step": 220
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.39464285714285713,
|
| 1561 |
+
"grad_norm": 0.2768295705318451,
|
| 1562 |
+
"learning_rate": 0.00013574790207974646,
|
| 1563 |
+
"loss": 1.3123,
|
| 1564 |
+
"step": 221
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.3964285714285714,
|
| 1568 |
+
"grad_norm": 0.2606358230113983,
|
| 1569 |
+
"learning_rate": 0.0001352138672882555,
|
| 1570 |
+
"loss": 1.4506,
|
| 1571 |
+
"step": 222
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.3982142857142857,
|
| 1575 |
+
"grad_norm": 0.24806426465511322,
|
| 1576 |
+
"learning_rate": 0.00013467868358438563,
|
| 1577 |
+
"loss": 1.7087,
|
| 1578 |
+
"step": 223
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.4,
|
| 1582 |
+
"grad_norm": 0.2664608061313629,
|
| 1583 |
+
"learning_rate": 0.00013414236842941644,
|
| 1584 |
+
"loss": 1.3124,
|
| 1585 |
+
"step": 224
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.4017857142857143,
|
| 1589 |
+
"grad_norm": 0.2661263346672058,
|
| 1590 |
+
"learning_rate": 0.00013360493932154302,
|
| 1591 |
+
"loss": 1.2174,
|
| 1592 |
+
"step": 225
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.4035714285714286,
|
| 1596 |
+
"grad_norm": 0.3460038900375366,
|
| 1597 |
+
"learning_rate": 0.00013306641379530514,
|
| 1598 |
+
"loss": 0.6889,
|
| 1599 |
+
"step": 226
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.40535714285714286,
|
| 1603 |
+
"grad_norm": 0.2929069995880127,
|
| 1604 |
+
"learning_rate": 0.000132526809421015,
|
| 1605 |
+
"loss": 0.9457,
|
| 1606 |
+
"step": 227
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.40714285714285714,
|
| 1610 |
+
"grad_norm": 0.3459819257259369,
|
| 1611 |
+
"learning_rate": 0.00013198614380418412,
|
| 1612 |
+
"loss": 1.2547,
|
| 1613 |
+
"step": 228
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.4089285714285714,
|
| 1617 |
+
"grad_norm": 0.30105313658714294,
|
| 1618 |
+
"learning_rate": 0.00013144443458494882,
|
| 1619 |
+
"loss": 0.957,
|
| 1620 |
+
"step": 229
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.4107142857142857,
|
| 1624 |
+
"grad_norm": 0.3461960256099701,
|
| 1625 |
+
"learning_rate": 0.00013090169943749476,
|
| 1626 |
+
"loss": 1.3146,
|
| 1627 |
+
"step": 230
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.4125,
|
| 1631 |
+
"grad_norm": 0.34542855620384216,
|
| 1632 |
+
"learning_rate": 0.00013035795606948023,
|
| 1633 |
+
"loss": 1.1128,
|
| 1634 |
+
"step": 231
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.4142857142857143,
|
| 1638 |
+
"grad_norm": 0.37605586647987366,
|
| 1639 |
+
"learning_rate": 0.00012981322222145846,
|
| 1640 |
+
"loss": 1.5095,
|
| 1641 |
+
"step": 232
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.4160714285714286,
|
| 1645 |
+
"grad_norm": 0.37267056107521057,
|
| 1646 |
+
"learning_rate": 0.00012926751566629875,
|
| 1647 |
+
"loss": 1.071,
|
| 1648 |
+
"step": 233
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.41785714285714287,
|
| 1652 |
+
"grad_norm": 0.3052172064781189,
|
| 1653 |
+
"learning_rate": 0.00012872085420860665,
|
| 1654 |
+
"loss": 1.3136,
|
| 1655 |
+
"step": 234
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.41964285714285715,
|
| 1659 |
+
"grad_norm": 0.36694592237472534,
|
| 1660 |
+
"learning_rate": 0.00012817325568414297,
|
| 1661 |
+
"loss": 1.2439,
|
| 1662 |
+
"step": 235
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.42142857142857143,
|
| 1666 |
+
"grad_norm": 0.36055245995521545,
|
| 1667 |
+
"learning_rate": 0.00012762473795924204,
|
| 1668 |
+
"loss": 1.1165,
|
| 1669 |
+
"step": 236
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.4232142857142857,
|
| 1673 |
+
"grad_norm": 0.3014545738697052,
|
| 1674 |
+
"learning_rate": 0.00012707531893022854,
|
| 1675 |
+
"loss": 1.5423,
|
| 1676 |
+
"step": 237
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.425,
|
| 1680 |
+
"grad_norm": 0.3208891749382019,
|
| 1681 |
+
"learning_rate": 0.00012652501652283377,
|
| 1682 |
+
"loss": 1.1813,
|
| 1683 |
+
"step": 238
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.42678571428571427,
|
| 1687 |
+
"grad_norm": 0.38703230023384094,
|
| 1688 |
+
"learning_rate": 0.00012597384869161084,
|
| 1689 |
+
"loss": 0.7706,
|
| 1690 |
+
"step": 239
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.42857142857142855,
|
| 1694 |
+
"grad_norm": 0.38256821036338806,
|
| 1695 |
+
"learning_rate": 0.00012542183341934872,
|
| 1696 |
+
"loss": 1.0565,
|
| 1697 |
+
"step": 240
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.4303571428571429,
|
| 1701 |
+
"grad_norm": 0.3555380702018738,
|
| 1702 |
+
"learning_rate": 0.0001248689887164855,
|
| 1703 |
+
"loss": 0.849,
|
| 1704 |
+
"step": 241
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.43214285714285716,
|
| 1708 |
+
"grad_norm": 0.3472703993320465,
|
| 1709 |
+
"learning_rate": 0.00012431533262052098,
|
| 1710 |
+
"loss": 1.3984,
|
| 1711 |
+
"step": 242
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.43392857142857144,
|
| 1715 |
+
"grad_norm": 0.3631349503993988,
|
| 1716 |
+
"learning_rate": 0.000123760883195428,
|
| 1717 |
+
"loss": 0.8955,
|
| 1718 |
+
"step": 243
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.4357142857142857,
|
| 1722 |
+
"grad_norm": 0.349295973777771,
|
| 1723 |
+
"learning_rate": 0.00012320565853106316,
|
| 1724 |
+
"loss": 0.8866,
|
| 1725 |
+
"step": 244
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.4375,
|
| 1729 |
+
"grad_norm": 0.33635953068733215,
|
| 1730 |
+
"learning_rate": 0.00012264967674257646,
|
| 1731 |
+
"loss": 1.2419,
|
| 1732 |
+
"step": 245
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.4392857142857143,
|
| 1736 |
+
"grad_norm": 0.3833181858062744,
|
| 1737 |
+
"learning_rate": 0.00012209295596982042,
|
| 1738 |
+
"loss": 1.5507,
|
| 1739 |
+
"step": 246
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.44107142857142856,
|
| 1743 |
+
"grad_norm": 0.3737214505672455,
|
| 1744 |
+
"learning_rate": 0.00012153551437675821,
|
| 1745 |
+
"loss": 1.4881,
|
| 1746 |
+
"step": 247
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.44285714285714284,
|
| 1750 |
+
"grad_norm": 0.4705282747745514,
|
| 1751 |
+
"learning_rate": 0.00012097737015087094,
|
| 1752 |
+
"loss": 1.4864,
|
| 1753 |
+
"step": 248
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.4446428571428571,
|
| 1757 |
+
"grad_norm": 0.39539188146591187,
|
| 1758 |
+
"learning_rate": 0.00012041854150256433,
|
| 1759 |
+
"loss": 1.7855,
|
| 1760 |
+
"step": 249
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.44642857142857145,
|
| 1764 |
+
"grad_norm": 0.7369075417518616,
|
| 1765 |
+
"learning_rate": 0.00011985904666457455,
|
| 1766 |
+
"loss": 2.01,
|
| 1767 |
+
"step": 250
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.44821428571428573,
|
| 1771 |
+
"grad_norm": 0.18146094679832458,
|
| 1772 |
+
"learning_rate": 0.00011929890389137337,
|
| 1773 |
+
"loss": 1.5898,
|
| 1774 |
+
"step": 251
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.45,
|
| 1778 |
+
"grad_norm": 0.21558880805969238,
|
| 1779 |
+
"learning_rate": 0.00011873813145857249,
|
| 1780 |
+
"loss": 1.5816,
|
| 1781 |
+
"step": 252
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.4517857142857143,
|
| 1785 |
+
"grad_norm": 0.19599275290966034,
|
| 1786 |
+
"learning_rate": 0.00011817674766232734,
|
| 1787 |
+
"loss": 1.6433,
|
| 1788 |
+
"step": 253
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.45357142857142857,
|
| 1792 |
+
"grad_norm": 0.22075910866260529,
|
| 1793 |
+
"learning_rate": 0.00011761477081874015,
|
| 1794 |
+
"loss": 1.6005,
|
| 1795 |
+
"step": 254
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.45535714285714285,
|
| 1799 |
+
"grad_norm": 0.19471955299377441,
|
| 1800 |
+
"learning_rate": 0.0001170522192632624,
|
| 1801 |
+
"loss": 1.7133,
|
| 1802 |
+
"step": 255
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.45714285714285713,
|
| 1806 |
+
"grad_norm": 0.19876879453659058,
|
| 1807 |
+
"learning_rate": 0.00011648911135009634,
|
| 1808 |
+
"loss": 1.5085,
|
| 1809 |
+
"step": 256
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.4589285714285714,
|
| 1813 |
+
"grad_norm": 0.20565317571163177,
|
| 1814 |
+
"learning_rate": 0.00011592546545159645,
|
| 1815 |
+
"loss": 1.7386,
|
| 1816 |
+
"step": 257
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.4607142857142857,
|
| 1820 |
+
"grad_norm": 0.24483506381511688,
|
| 1821 |
+
"learning_rate": 0.00011536129995766996,
|
| 1822 |
+
"loss": 1.7162,
|
| 1823 |
+
"step": 258
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.4625,
|
| 1827 |
+
"grad_norm": 0.21543823182582855,
|
| 1828 |
+
"learning_rate": 0.00011479663327517667,
|
| 1829 |
+
"loss": 1.6966,
|
| 1830 |
+
"step": 259
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.4642857142857143,
|
| 1834 |
+
"grad_norm": 0.2661048471927643,
|
| 1835 |
+
"learning_rate": 0.00011423148382732853,
|
| 1836 |
+
"loss": 1.8821,
|
| 1837 |
+
"step": 260
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.4660714285714286,
|
| 1841 |
+
"grad_norm": 0.24292460083961487,
|
| 1842 |
+
"learning_rate": 0.00011366587005308858,
|
| 1843 |
+
"loss": 1.7085,
|
| 1844 |
+
"step": 261
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.46785714285714286,
|
| 1848 |
+
"grad_norm": 0.216167613863945,
|
| 1849 |
+
"learning_rate": 0.0001130998104065693,
|
| 1850 |
+
"loss": 1.7298,
|
| 1851 |
+
"step": 262
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.46964285714285714,
|
| 1855 |
+
"grad_norm": 0.2111697793006897,
|
| 1856 |
+
"learning_rate": 0.00011253332335643043,
|
| 1857 |
+
"loss": 1.8098,
|
| 1858 |
+
"step": 263
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.4714285714285714,
|
| 1862 |
+
"grad_norm": 0.23981061577796936,
|
| 1863 |
+
"learning_rate": 0.00011196642738527659,
|
| 1864 |
+
"loss": 1.7026,
|
| 1865 |
+
"step": 264
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.4732142857142857,
|
| 1869 |
+
"grad_norm": 0.2623251676559448,
|
| 1870 |
+
"learning_rate": 0.00011139914098905406,
|
| 1871 |
+
"loss": 1.7894,
|
| 1872 |
+
"step": 265
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.475,
|
| 1876 |
+
"grad_norm": 0.2482486367225647,
|
| 1877 |
+
"learning_rate": 0.00011083148267644747,
|
| 1878 |
+
"loss": 1.9019,
|
| 1879 |
+
"step": 266
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.4767857142857143,
|
| 1883 |
+
"grad_norm": 0.238911435008049,
|
| 1884 |
+
"learning_rate": 0.00011026347096827578,
|
| 1885 |
+
"loss": 1.6809,
|
| 1886 |
+
"step": 267
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.4785714285714286,
|
| 1890 |
+
"grad_norm": 0.24704696238040924,
|
| 1891 |
+
"learning_rate": 0.00010969512439688816,
|
| 1892 |
+
"loss": 1.6607,
|
| 1893 |
+
"step": 268
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.48035714285714287,
|
| 1897 |
+
"grad_norm": 0.25105100870132446,
|
| 1898 |
+
"learning_rate": 0.00010912646150555919,
|
| 1899 |
+
"loss": 1.5895,
|
| 1900 |
+
"step": 269
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.48214285714285715,
|
| 1904 |
+
"grad_norm": 0.2849842607975006,
|
| 1905 |
+
"learning_rate": 0.00010855750084788398,
|
| 1906 |
+
"loss": 2.0812,
|
| 1907 |
+
"step": 270
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.48392857142857143,
|
| 1911 |
+
"grad_norm": 0.2599342167377472,
|
| 1912 |
+
"learning_rate": 0.00010798826098717276,
|
| 1913 |
+
"loss": 1.1569,
|
| 1914 |
+
"step": 271
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.4857142857142857,
|
| 1918 |
+
"grad_norm": 0.28037258982658386,
|
| 1919 |
+
"learning_rate": 0.00010741876049584523,
|
| 1920 |
+
"loss": 1.1928,
|
| 1921 |
+
"step": 272
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.4875,
|
| 1925 |
+
"grad_norm": 0.29920563101768494,
|
| 1926 |
+
"learning_rate": 0.00010684901795482456,
|
| 1927 |
+
"loss": 1.2244,
|
| 1928 |
+
"step": 273
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.48928571428571427,
|
| 1932 |
+
"grad_norm": 0.2799164354801178,
|
| 1933 |
+
"learning_rate": 0.00010627905195293135,
|
| 1934 |
+
"loss": 1.4455,
|
| 1935 |
+
"step": 274
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.49107142857142855,
|
| 1939 |
+
"grad_norm": 0.23873603343963623,
|
| 1940 |
+
"learning_rate": 0.00010570888108627681,
|
| 1941 |
+
"loss": 0.852,
|
| 1942 |
+
"step": 275
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.4928571428571429,
|
| 1946 |
+
"grad_norm": 0.2817741632461548,
|
| 1947 |
+
"learning_rate": 0.00010513852395765631,
|
| 1948 |
+
"loss": 1.3203,
|
| 1949 |
+
"step": 276
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.49464285714285716,
|
| 1953 |
+
"grad_norm": 0.27295514941215515,
|
| 1954 |
+
"learning_rate": 0.00010456799917594233,
|
| 1955 |
+
"loss": 0.749,
|
| 1956 |
+
"step": 277
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.49642857142857144,
|
| 1960 |
+
"grad_norm": 0.30728739500045776,
|
| 1961 |
+
"learning_rate": 0.00010399732535547734,
|
| 1962 |
+
"loss": 1.0083,
|
| 1963 |
+
"step": 278
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.4982142857142857,
|
| 1967 |
+
"grad_norm": 0.32001444697380066,
|
| 1968 |
+
"learning_rate": 0.00010342652111546635,
|
| 1969 |
+
"loss": 1.573,
|
| 1970 |
+
"step": 279
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.5,
|
| 1974 |
+
"grad_norm": 0.3400101065635681,
|
| 1975 |
+
"learning_rate": 0.00010285560507936961,
|
| 1976 |
+
"loss": 1.0757,
|
| 1977 |
+
"step": 280
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.5,
|
| 1981 |
+
"eval_loss": 1.4439584016799927,
|
| 1982 |
+
"eval_runtime": 13.9741,
|
| 1983 |
+
"eval_samples_per_second": 16.888,
|
| 1984 |
+
"eval_steps_per_second": 8.444,
|
| 1985 |
+
"step": 280
|
| 1986 |
}
|
| 1987 |
],
|
| 1988 |
"logging_steps": 1,
|
|
|
|
| 2002 |
"attributes": {}
|
| 2003 |
}
|
| 2004 |
},
|
| 2005 |
+
"total_flos": 9.129139501635994e+16,
|
| 2006 |
"train_batch_size": 2,
|
| 2007 |
"trial_name": null,
|
| 2008 |
"trial_params": null
|